The Expression of the ER Stress Sensor GRP78/Bip Is a Target of R-CHOP and Bortezomib Treatments in DLBCL with Prognostic Value. A Rationale for the Use of Proteasome Inhibitors in DLBCL Patients.

The endoplasmic reticulum (ER) stress response is an adaptive signaling pathway that controls cell survival. The activation of the transcription factor Xbp1 is a main event in this response and we have previously shown that Xbp1 activation in DLBCL is associated with aggressive clinical course (Balagué et al. Am J Pathol 2009, 174(6):2337-46). GRP78/Bip is a molecular chaperone that senses ER homeostasis and initiates the ER stress response. The expression of GRP78/Bip in DLBCL has never been addressed before. DLBCL patients are treated with standard doxorubicin-based chemotherapy such as CHOP. Since the introduction of rituximab, no other therapies have shown greater benefit in these patients. The ER stress response may be altered by conventional chemotherapy and it is well known that proteasome inhibition with bortezomib disrupts this response in myeloma. Whether Bip is affected in DLBCL treated with R-CHOP or bortezomib is unknown. Recent evidences suggest that the addition of bortezomib to standard chemotherapy would improve the survival of patients with the DLBCL preferentially of the ABC subtype (Duneleavy et al. Blood 2009, 113(24):6069-76) but the implication of the ER stress in this combination therapy remains unknown.

We analyze the mRNA and protein expression of Bip in DLBCL cell lines (OCI-Ly8, SUHDL4, SUDHL6 and SUDHL16) treated with Bortezomib, R-CHOP or their combination. Moreover, we also evaluated the effect of Bip silencing in the response to these treatments by using siRNA assay. Cell viability was analyzed by Annexin V. We also analyze the expression of Bip in 138 DLBCL patients by immunohistochemistry and in 63 of them by mRNA gene expression. Clinical data and follow up were available in 52 patients with a mean follow up of 2.9 years (range 0.02 to 6.7 years).

All cell lines responded to R-CHOP treatment, with a decrease in cell viability ranging from 20% observed in OCI-LY8 cells to 45% in SUDHL6 cells. Moreover, in parallel with this response we detected a marked decrease in Bip expression both by protein and qPCR analysis. Bortezomib alone was less effective than R-CHOP, with 25% decrease in cell viability in the most sensitive SUDHL6 cells and less than 1% decrease in cell viability in the most resistant OCI-LY8 cells. Bortezomib increased both BiP mRNA and protein expression. The combination of bortezomib plus R-CHOP induces higher rates of cell death in all cell lines ranging between 35% decrease in cell viability in OCI-LY8 cells to 53.7% in SUDHL16 cells. This combination therapy led to an increase of Bip mRNA and protein expression but at much less extent than bortezomib alone. To confirm that BiP plays an antiapoptotic role in DLBCL we performed a siRNA assay for Bip in OCI-LY8 and SUDHL16 cell lines, corresponding to the most resistant and sensitive cell lines. After siRNA transfection, both cell lines reduced 60% to 80% Bip mRNA expression and became sensitive to bortezomib alone and more sensitive to the combination therapy. Bip expression was observed in 96 (69.5%) of newly diagnosed DLBCL tumors independently of Xbp1 activation and ABC subtype. Moreover high Bip mRNA expression (3-4 quartile) was predictive of worse survival (median overall survival 3.34 vs 1.9 years, p=0.048).

The ER-stress sensor Bip is expressed in DLBCL cell lines and primary tumors and it plays an important prosurvival role. Moreover Bip expression is a target of R-CHOP and bortezomib treatments being responsible for the primary resistance to bortezomib. In addition, the combination of R-CHOP plus bortezomib reduced Bip expression and decreased cell survival providing a rationale for the combination therapy in the clinical settings. Furthermore, high Bip expression is an adverse prognostic factor in newly diagnosed DLBCL patients treated with R-CHOP and may be used to select patients that would benefit from the addition of bortezomib to the standard chemotherapy.

No relevant conflicts of interest to declare.